“Many of the funded researchers kept an affiliation with Germany even after the funding had ended”

What was the central research question of your analysis and what methodology did you use?

Johann Mouton: Together with DAAD, we set out to investigate the academic career tracks of doctoral graduates who had received funding from DAAD. Our starting point were bibliometric records of the graduates, meaning publications produced during and after completion of a doctoral programme. Based on those indications of academic work, we aimed at identifying patterns – how active were the researchers, did they change their affiliations, and how much were they involved in collaborations with researchers from Germany?

To begin with, finding and identifying the publications of the DAAD-funded researchers was easier said than done. We started with a list of about 4,800 DAAD-funded researchers and had to identify the equivalents of these researchers in other data records. First, we matched the names with authors in a publication database – Web of Science. Here, we got help from an algorithm. Results were validated manually, also taking into account information from other platforms, such as ORCID, a database that provides persistent digital identifiers – so-called ORCID iDs – for researchers. With an ORCID iD, a researcher can be distinguished from every other researcher in the world. The information we got from this process was augmented by including author records from other publication databases. In the end, we had a rich dataset that allowed us to study the publication output, collaboration networks and affiliations of the DAAD-funded researchers quite extensively.

Can you summarize the main findings of the analysis for us?

Jonathan Dudek: Sure. To begin with, we learned quite a bit in the process of conducting this analysis. When identifying the outputs of the funded researchers, we had to face the challenge of ambiguous author names. It turned out that the ORCID iDs were of great help in this process, at least when they existed. However, only about 48 percent of the researchers in our dataset could be connected with such an identifier. Even researchers for whom we could find publications did not always have an ORCID iD. This shows that there is clearly room for improvement, so that individual researchers can properly be linked to their outputs.

But let’s talk about the findings of the study. We can say that the share of researchers affiliated with Germany increased during the funding period. While that could be expected, we also found that many of the funded researchers kept an affiliation with Germany even after the funding had ended. We think that this may indicate that the researchers have built up academic relationships in Germany, even beyond the funding. It may also suggest that the funding contributes to lasting, sustainable research networks.

Rodrigo Costas: Related to this, our results hint to an interesting aspect regarding the German collaborators of the funded researchers: They showed increasing collaborations with the countries the funded researchers came from – even after the funding period. Most interestingly, this was also true when we excluded collaborations including the DAAD-funded scholars. This result supports the idea that expanded collaboration networks are developed before, during, and after the funding.

Another result worth highlighting is the diversity of the origins of the funded researchers with publications, both geographically and in terms of fields of research. Still, we found a couple of countries that stand out, with Russia, Pakistan, Egypt, and the United States leading. Regarding fields of research, the biomedical and health sciences are related to the biggest part of the publishing researchers: two thirds. In contrast to that, the social sciences and humanities account for only about ten percent, and mathematics and computer science for about nine percent.

In your view, what are the most important scientific and practical conclusions that result from your findings?

Mouton: Generally speaking, it was good to see that a bibliometric analysis can be quite insightful for a study of the development of funded researchers. This goes far beyond just counting publications or citations, but includes the whole bandwidth of topics, geographical connections, and collaboration networks.

Dudek: In a very practical sense, our study shows the importance of using identifiers such as the ORCID iD to reliably capture and connect research activities and individual funding. This has implications for the researchers themselves, who can make sure that their works will be counted properly. But also for a funding organisation like the DAAD, requiring an ORCID iD from funded researchers is essential for studying the publication activities of funded researchers.

Costas: Finally, this study has created a number of ideas for future research. Apparently, funding can be related to lasting research collaboration networks. It would be very interesting to look into how exactly that works, for example how funding through DAAD may be a contributing factor in the creation and sustaining of affiliations with Germany. Another very interesting aspect would be to look closer into the beneficial role of the funding for the German collaborators and like that, the potential spill-over effects of funding such as that from the DAAD.